trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.90727969348659,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "learning_rate": 9.999693232713664e-06,
      "loss": 0.0295,
      "step": 150
    },
    {
      "epoch": 1.84,
      "learning_rate": 9.998772930854655e-06,
      "loss": 0.0272,
      "step": 300
    },
    {
      "epoch": 2.45,
      "eval_loss": 0.23613713681697845,
      "eval_runtime": 411.8281,
      "eval_samples_per_second": 11.311,
      "eval_steps_per_second": 1.416,
      "eval_wer": 0.15606540535913022,
      "step": 400
    },
    {
      "epoch": 2.76,
      "learning_rate": 9.997852628995645e-06,
      "loss": 0.0272,
      "step": 450
    },
    {
      "epoch": 3.68,
      "learning_rate": 9.996932327136635e-06,
      "loss": 0.0268,
      "step": 600
    },
    {
      "epoch": 4.6,
      "learning_rate": 9.996012025277625e-06,
      "loss": 0.0261,
      "step": 750
    },
    {
      "epoch": 4.91,
      "eval_loss": 0.2315920889377594,
      "eval_runtime": 257.9698,
      "eval_samples_per_second": 18.056,
      "eval_steps_per_second": 2.26,
      "eval_wer": 0.15478126872699255,
      "step": 800
    }
  ],
  "max_steps": 1630000,
  "num_train_epochs": 10000,
  "total_flos": 1.8155123522141764e+19,
  "trial_name": null,
  "trial_params": null
}